|Home | About | Journals | Submit | Contact Us | Français|
Patients with amyotrophic lateral sclerosis (ALS) have evidence of chronic inflammation demonstrated by infiltration of the gray matter by inflammatory macrophages, IL17A-positive T cells, and mast cells. Increased serum levels of IL6 and IL17A have been detected in sporadic ALS (sALS) patients when compared to healthy controls. Herein we investigate, in peripheral blood mononuclear cells (PBMCs), the baseline transcription of genes associated with inflammation in sALS and control subjects and the impact of the IL6 receptor (IL6R) antibody (tocilizumab) on the transcription and/or secretion of inflammation factors (e.g. cytokines) stimulated by the apo-G37R superoxide dismutase (SOD1) mutant. At baseline, PBMCs of four sALS patients (Group 1) showed significantly increased expression of TLR2 and CD14; ALOX5, PTGS2 and MMP1; IL1α, IL1β, IL6, IL36G, IL8 and TNF; CCL3, CCL20, CXCL2, CXCL3 and CXCL5. In four other sALS patients (Group 2), most of the genes just mentioned were expressed at near control levels and a significant decrease in the expression of PPARG, PPARA, RARG, HDAC4 and KAT2B; IL6R, IL6ST and ADAM17; TNFRSF11A; MGAT2 and MGAT3; PLCG1; CXCL3 were detected. Apo-G37R SOD1 up regulated the transcription of cytokines (e.g. IL1α/β, IL6, IL8, IL36G), chemokines (e.g. CCL20; CXCL3, CXCL5), and enzymes (e.g. PTGS2 and MMP1). In vitro, tocilizumab down regulated the transcription of many inflammatory cytokines, chemokines, enzymes, and receptors, which were up regulated by pathogenic forms of SOD1. Tocilizumab also reduced the secretion of the pro-inflammatory cytokines IL1β, IL6, TNFα, GM-CSF, IFNγ, and IL17A by Group 1 PBMCs. Finally, sALS patients had significantly higher concentrations of IL6, sIL6R and C-reactive protein in the cerebrospinal fluid when compared to AD patients. This pilot study demonstrates that in vitro tocilizumab suppresses many factors that drive inflammation in sALS patients, with possible increased efficacy in Group 1 ALS patients.
Neuronal degeneration in the amyotrophic lateral sclerosis (ALS) spinal cord is associated with chronic inflammation and is marked by infiltrating IL1β-, IL6-, and TNFα-positive macrophages/microglia [1,2], as well as IL17A-positive CD8 and mast cells . The inflammatory cytokines IL1β, IL6, TNFα, GM-CSF, and the bi-functional cytokine IL10, are all induced in the peripheral blood mononuclear cells (PBMCs) of ALS patients by stimulation with demetallated (Apo) or fibrillar wild-type SOD1 . Mutant forms of SOD1 appear to activate the inflammation in monocytes/macrophages through activation of cyclooxygenase-2 (PTGS2 or COX-2) and caspase-1 . Inflammatory macrophages, expressing IL6 and TNF-α, have been observed to phagocytize both apoptotic and non-apoptotic neurons in the spinal cord , suggesting a potential immune mechanism that promotes neuronal death in ALS. Systemic inflammation has also been observed in early stages of the disease in ALS mouse models [6,7].
Interleukin-6, together with the cytokine TGFβ, are well-known to promote the development of Th17 cells , which support chronic inflammation in autoimmune diseases . IL6 is a bifunctional cytokine with both pro-inflammatory and anti-inflammatory activities: (a) “classical” signaling through the membrane-bound IL6 receptor (IL6R) for neuroprotective activities, and (b) “trans-signaling” by a complex formed between a soluble IL6 receptor (sIL6R) and IL6 for pro-inflammatory activities. The sIL6R-IL6 complex allows IL6 signaling in cells lacking IL6R by binding to the signaling IL6 co-receptor gp130 (gp130R) . Soluble IL6R is present in serum, urine, synovial fluid, and cerebrospinal fluid (CSF) of normal subjects and is increased in subjects with autoimmune diseases . IL6/sIL6R trans-signaling has been shown to stimulate chronic inflammation in rheumatoid joints .
The role of IL6 signaling in neurological diseases is not clear . For example, Alzheimer’s disease (AD) patients had high plasma levels of IL6, TNFα, and IL1β  and an increased CSF level of IL6 . The CSF levels of sIL6R were found to be decreased in one study , but in another study equally elevated CSF concentrations of sIL6R (mean 1,000 pg/ml ) in both AD patients and normal subjects were observed . Accordingly, the roles of the IL6/sIL6R trans-signaling in the neurodegenerative diseases may vary in different stages of the disease, and may depend upon the ratios of free IL6 and sIL6R or other factors specific to each disease .
The IL6R antibody called tocilizumab (ActemraR) inhibits IL6 signaling through both IL6R and sIL6R. Tocilizumab has shown favorable long-term effects in patients with rheumatoid arthritis [18-20], and is under study in patients with Castleman’s disease, juvenile rheumatoid arthritis, and inflammatory bowel disease  . In this study, we investigated PBMCs of ALS patients and controls regarding changes in the transcription of inflammatory cytokines, chemokines and their cognate receptors, as well as other genes; and the effect of tocilizumab on the transcription and/or secretion of cytokines and chemokines. We also showed the evidence of IL6-related inflammation in the ALS spinal cord and the cerebrospinal fluid (CSF) by testing the IL6, sIL6R, C-reactive protein (CRP) and IL1 receptor antagonist (IL1RA) levels.
Eight patients with the sporadic ALS (sALS) diagnosis, 4 normal controls and one unaffected twin of an sALS patient were enrolled into the study according to the UCLA IRB approved protocol. PBMC and macrophage cultures were done as previously described.
PBMCs were separated from heparin-anticoagulated blood by Ficoll-Hypaque gradient centrifugation. From each subject five million PBMCs per tube (treatment) were cultured overnight with or without 2 μg/ml of SOD1 protein. After 16-18 hours, supernatants were collected from each tube and frozen immediately for future batch testing of cytokines. The supernatant of each patient was tested with R&D Systems High Sensitivity Human Inflammation Multiplex-Kit – Pre-mixed 12- human cytokines using Luminex platform of Bio-RAD BioPlex 200 dual laser, flow-based sorting and detection analyzer. This multi-plex kit simultaneously quantified supernatant concentrations of human IL-1ß, interleukin-2 (IL2), interleukin-4 (IL4), interleukin-5 (IL5), IL6, interleukin-7 (IL7), interleukin-8 (IL8), interleukin-10 (IL10), interleukin-12 (IL12), IL13, interferon-γ (γ), granulocyte-macrophage colony stimulating factor (GM-CSF), and tumor necrosis factor-α (TNFα). The results are presented in pg/ml.
Following isolation by the Ficoll-Hypaque technique, approximately 4.0 x 106 PBMCs were incubated overnight in IMDM medium alone (baseline), IMDM with 2 μg/ml SOD1, or IMDM with tocilizumab (2 μg/ml) and SOD1. Cells were pelleted by centrifugation and resuspended in RNA preserve medium (Qiagen, Valencia, CA). Total RNA was extracted using the RNeasy Mini-prep (+ deoxyribonuclease step) kit and, as needed, concentrated and cleaned up (Qiagen RNeasy MinElute Cleanup kit) on the day the array was plated. cDNA was then prepared from 200 ng of total RNA using the RT2 First Strand Kit and was added to RT-PCR reagent SYBR Green Master Mix according to manufacturer protocol (Qiagen, Valencia, CA, USA). Ten microliters of the mixture was added to each of the 384-wells of the RT2 Profiler inflammatory and autoimmune gene array PAHS-077G (Qiagen, Valencia, CA, USA). The RT-PCR reaction was performed on the Roche LightCycler 480. Data was processed by the ΔΔCt method using proprietary tools supplied by SABiosciences PCR Data Analysis Web Portal (Qiagen, Valencia, CA, USA). Each custom array included control wells of (a) the threshold cycle values for genomic DNA contamination, (b) inhibition of reverse transcription, (c) a positive PCR control, and (d) three housekeeping genes. The fold regulation for a given gene was calculated by comparing a control to a treatment group and using the 2 (-ΔΔCt) normalized with housekeeping genes and other genes on the array with recorded Ct values within 0.5 cycles across the control and treatment groups.
50,000 PBMCs were cultivated for 1, 5 and 8 days in the presence of aggregated SOD1 (2 μg/ml) and ALS macrophages in 8-well chamber slides. The supernatant fluids were removed for IL17A testing and the cells were stained by immunofluorescence with anti-IL17A.
The differences between the groups were tested by t-test following the Levine test for equality of variances. Correlation between cytokines was tested by Pearson’s r and Spearman’s rho.
We previously showed by Affymetrix microarray hybridization  and qRT-PCR  that wild-type SOD1 in the Apo (demetallated) or fibrillar forms induced significantly greater transcription of inflammatory cytokines and chemokines in PBMCs of sALS patients when compared to control subjects. Here we have investigated the degree of baseline (i.e. background) inflammation in PBMCs of sporadic ALS patients and controls by qRT- PCR analysis of transcription of ninety genes associated with inflammatory signaling. The PCR results suggested two groups in the study population (n=8) based on the fold regulation change in the mRNA levels of IL1 and IL6 mRNA expression when compared to the control subjects (n=4): Group 1 patients (n=4, Table 1) showed increased expression and Group 2 patients (n=4, Table 1) showed decreased or unaltered expression of IL1 and IL6.
Other genes significantly up regulated in the Group 1 PBMCs were TLR2, CD14, IL1, IL8, TNF, ALOX5, CCL3, CCL20, CXCL2, CXCL3, and CXCL5 (Figure 1A). Alternatively, significant down regulation of PPARg, PPARα, RARg, HDAC4, KAT2B, IL6R, IL6ST, TNFRSF11A, MGAT2, MGAT3, ADAM17, PLCG1, TSC1, TSC2 and CXCL3 was observed in the Group 2 PBMCs (Figure 1A). When compared to each other, Group 1 patients showed significantly higher mRNA levels of cytokines (e.g. IL1 and IL23A), chemokines (e.g. CXCL3 and CCL20), matrix metalloproteinases (e.g. MMP1 and MMP14), and TLR2 than Group 2 patients (Figure 1B).
RT-PCR testing of PBMCs of a Group 1 ALS patient and her unaffected identical twin revealed that only the ALS patient had increased transcription of IL1α, IL6, MMP1, and CXCL3 (Figure 2A); however, both showed increased transcription of CXCL1, CXCL5, IL1β, and CCL20 when compared to controls (Figures 2B and and2C2C).
To test the effect of tocilizumab on IL-6/sIL6 R signaling and the downstream inflammatory genes, we first stimulated PBMCs with apo-G37R SOD1, a pathogenic form of SOD1 associated with familial ALS. As shown previously , apo and fibrillar forms of wild type SOD1 or apo-G37R SOD-1 stimulate IL1 and IL6 and a variety of chemokines (e.g. CCL19). Here apo-G37R SOD1 stimulated the transcription of the genes already up regulated at baseline in Group 1 and those that were similar to control subjects at baseline in Group 2 (Table 2). For the most part a more pronounced induction of cytokine and chemokine mRNA expression by apo-G37R was observed in Group 1 PBMCs in this pilot study. We are extending these observations in a larger sample of ALS patients.
Tocilizumab (10 μg/ml) broadly inhibited the transcription of many cytokines (e.g. IL1 and IL6), some chemokines (e.g. CCL19 and CCL20) and other factors previously shown to be up regulated by apo-G37R SOD1 in sALS in comparison to control PBMCs (e.g. TFPI2). Tocilizumab had greater inhibitory potency in Group 1 than Group 2 PBMCs, as shown by the stronger return of the mRNA expression level of these genes to baseline level in Group 1 (Table 2). In both ALS groups tocilizumab up regulated the expression of CXCL9 and had regulatory effects on other genes (e.g. CD14, MGATs, ALOXs, MMPs etc., Table 2).
In addition to transcriptional effects, tocilizumab inhibited the secretion of IL1β, IL6, IL10, GM-CSF and TNFα, all of which were increased by stimulation of ALS PBMCs by apo G37R SOD-1 (Figure 3). Because IL17A is the cytokine involved in chronic autoimmune diseases and has been shown to be intermittently elevated in ALS patients , we tested the effect tocilizumab on the expression and secretion of IL17A. As shown previously, stimulation of PBMCs from a control subject co-cultured with macrophages from an sALS patient and with apo-G37R SOD1 elicits production of IL17A . Here we induced cellular expression of IL17A (Figure 4A) and secretion of IL17A (Figure 4B) when the macrophages of an ALS Group 1 patient were co-cultured with control PBMCs. Both the expression and the secretion of IL17A were attenuated by tocilizumab (Figure 4).
The ALS spinal cord is infiltrated by macrophages expressing IL6 and TNFα and by IL17A positive T cells and mast cells in the grey matter [3,4]. To evaluate the degree of inflammation in CSF, we measured in the CSF the levels of the inflammatory cytokines and receptors IL6, sIL6R, C-reactive protein and IL1RA, and compared these to the CSF levels of Alzheimer disease (AD) patients. IL6 and sIL6R were found in the CSF of ALS patients in significantly higher concentrations than in the CSF of AD patients (P = 0.008) (Figure 5A). Increased sIL6R correlated with IL6 (r=0.588, P= 0.057) and IL1RA correlated with CRP (r= 0.644, P=0.019, Figure 5B).
The inflammation in the ALS spinal cord and brain has been in the forefront of discussions regarding ALS pathogenesis and therapy for at least 20 years . Here we provide additional evidence of inflammation in the PBMCs and the CSF of ALS patients, although not all ALS patients show signs of systemic inflammation. Therefore, we separated the patients into two groups: Group 1 with high transcriptional activation and Group 2 with minor or no transcriptional activation with respect to controls. Our results in this pilot study do not clarify whether these groups reflect patients in different stages of the disease or indicate different course in patients belonging to one or the other group.
The important therapeutic objective was to investigate the effect of tocilizumab on inflammation in ALS patients. We demonstrated that tocilizumab attenuates the expression and secretion of inflammatory cytokines and chemokines in ALS PBMCs, more so in Group 1 compared to Group 2 patients.
The results of transcriptional testing showed crucial importance of IL1, IL6, MMP1, and certain chemokines in the progression of ALS. A comparison of a pair of twins, one diagnosed with ALS and the other unaffected, indicates similarities in the up regulation of the chemokines CCL20, CXCL1 and CXCL5 and the cytokine IL1β when compared to controls. However, the ALS twin showed increased expression of MMP1, CXCL3, IL1α, and IL6 in comparison to her unaffected twin. MMP1, CXCL3, IL1α, and IL6 were up regulated in ALS PBMCs following stimulation with aggregated SOD1 in both sALS groups, as previously shown with other sALS patients by RT PCR  and by microarray . Collectively, the RT-PCR results suggest that Group 1, but less so Group 2, patients have indicators of systemic inflammation. However, when presented with pathogenic SOD-1, a robust increase in the expression of chemokines, cytokines and activation of macrophages is observed in both Group 1 and Group 2 patients.
Tocilizumab had powerful effects on inflammation in ALS PBMCs stimulated by apo-G37R SOD1. In PBMCs of patients, tocilizumab inhibited both the transcription and secretion of cytokines and chemokines induced by apo-G37R. Of note, IL1, IL6, IL36G, IL8, IL23A, TNF, PTGS2, MMP1, TFPI2, CCL7, CCL19, CCL20, CXCL1, CXCL2, CXCL3 and CXCL5 were all down regulated by tocilizumab in both ALS patient groups. Tocilizumab also inhibited the expression and secretion of the cytokine IL17A induced by ALS macrophages in co-culture with healthy PBMCs. Thus macrophages may contribute to the chronic nature of inflammation in ALS patients [3,4] and the level of inflammation could be regulated by tocilizumab. The significance of inflammation in the ALS central nervous system was highlighted by demonstration of very high levels of IL6, sIL6 receptor and C-reactive protein in the CSF of ALS patients.
In conclusion, tocilizumab showed excellent activity against inflammatory activation in PBMCs of sALS patients in vitro. Further in vitro and in vivo studies are warranted to determine the anti-inflammatory effects of tozilizumab and potential clinical benefits of tocilizumab in sALS patients.
This investigation was supported by a grant from Genentech to M.F. and by NIH (NINDS) grant P01 NS049134 supporting M.C. We thank Nicole Mahanian and Rachel Weitzman for assistance with preparation of the manuscript. Immunoassays were carried out in the facilities of the UCLA AIDS Institute, which were supported, in part, by funds from the James B. Pendleton Charitable Trust and the McCarthy Family Foundation, and by NIH grant AI-028697: UCLA Center for AIDS Research (CFAR). We thank the National Neurological AIDS Bank , MH083500, and the UCLA Mary E. EASton ADRC Tissue brain bank, P50AG16570, for providing the spinal fluids from ALS and AD patients.
The authors have no conflict of interest.